[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: FW: [lead-webtb] Perl 5.8.4
- Subject: RE: FW: [lead-webtb] Perl 5.8.4
- Date: Mon, 29 Nov 2004 10:30:33 -0700 (MST)
Hey Robb,
That's an intersting theory that I'd be willing to believe because you
obviously tracked the problem down to the select satement and there seems
to be no other mention of a select bug solution(at least in what I've seen
so far) in the perl support info.
However, I'm a little uncomfortable telling the folks in LEAD that it is
the bug, because they specify Win32 which is ONLY a Windows deal, and they
specifically support Linux as a different line item. So I've just
responded by saying that we don't know a bug number and will leave it at
that. Michaels's unwillingness to support Perl to the level of LEAD's
need is much more the source of the problem if you ask me (and of course I
never said that! ;-)
Thanks for your help!
Tom.
On Mon, 29 Nov 2004, Robb Kambic wrote:
>
> it was the bug because it fixed the problem. if the bug appear on an
> intel based machines, ie windows or linux then it's possible that the os's
> used the same underlining hardware routines for the select() function. my
> guess is that if it appeared on windows then it would appear on linux.
> that being said, i don't keep up on that levels of os's.
>
> robb...
>
>
>
>
>
> On Mon, 22 Nov 2004, Tom Baltzer wrote:
>
> > Hey Robb,
> >
> > I see you're out for the week so I'll just pop this into your e-mail for
> > discussion at your return.
> >
> > I see that the text in the delta you point to is identical to the text I
> > pointed you to that indicates it was specifically for the Win32 platform....
> > are you sure that was the bug?
> >
> > Thanks,
> > Tom.
> >
> > Tom Baltzer
> > Software Engineer
> > UCAR Office of Programs
> > Unidata Program Center
> > P.O. Box 3000
> > Boulder, CO 80307-3000
> > address@hidden
> > 303-497-8637
> >
> >
> > > -----Original Message-----
> > > From: Robb Kambic [mailto:address@hidden]
> > > Sent: Friday, November 19, 2004 1:55 PM
> > > To: address@hidden; Tom Baltzer
> > > Cc: decoders
> > > Subject: Re: FW: [lead-webtb] Perl 5.8.4
> > >
> > > regarding the select() bug in 5.8.0
> > >
> > > ok, i can't find a bug number per se, but i did find a perl581delta - what
> > > is new for perl v5.8.1 that describes the select bug at:
> > >
> > > http://www.enstimac.fr/Perl/allpod/en-5.8.5/perl581delta.pod
> > >
> > > search for select a couple of times.
> > >
> > > i can't believe that no new perl rpm has been created for redhat, latest
> > > perl release is already 5.8.6 if michael has some manner of having linux
> > > folks making a new rpm, i would appreciate it as other would too. i think
> > > the fedora releases have perl version v5.8.3 though.
> > >
> > > robb...
> > >
> > > On Fri, 19 Nov 2004, Tom Baltzer wrote:
> > >
> > > > Hey Robb,
> > > >
> > > > I thought I'd remembered you identifying this bug, but can't seem to
> > > come by
> > > > that information in my e-mail archives. I did find that you wrote:
> > > >
> > > > The four-argument form of select() did not preserve $! (errno) properly
> > > when
> > > > there were errors in the underlying call. This is now fixed. version
> > > 5.8.1
> > > >
> > > > So I googled it up and found that this fix refers to Win32 architecture
> > > -
> > > > see: http://search.cpan.org/dist/perl/pod/perl581delta.pod
> > > >
> > > > Are you sure that was the problem? Do you know the perl bug number as
> > > > Michael asks for?
> > > >
> > > > Thanks,
> > > > Tom.
> > > >
> > > > Tom Baltzer
> > > > Software Engineer
> > > > UCAR Office of Programs
> > > > Unidata Program Center
> > > > P.O. Box 3000
> > > > Boulder, CO 80307-3000
> > > > address@hidden
> > > > 303-497-8637
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: address@hidden [mailto:address@hidden]
> > > On
> > > > Behalf Of Michael McEniry
> > > > Sent: Friday, November 19, 2004 11:31 AM
> > > > Cc: 'address@hidden'
> > > > Subject: Re: [lead-webtb] Perl 5.8.4
> > > >
> > > > LEAD - Web and Grid Test Bed Group Mailing List
> > > >
> > > > Marcus A Christie wrote:
> > > > > It's been a few weeks, but a while ago I believe it was Tom who said
> > > > > that we could install perl 5.8.4 to solve the runaway decoder problem
> > > > > (I've observed this with at least metar2nc and ua2nc). However, since
> > > > > the most up to date perl RPM from Red Hat for RHEL 3.0 isn't 5.8.4,
> > > that
> > > > > upgrade path isn't available. What have others done? Created a
> > > > > separate local manual installation of perl?
> > > >
> > > > I refuse to maintain a throwaway installation of something as cumbersome
> > > > as perl. If Unidata has identified a known bug (that has already been
> > > > fixed!!!) in perl, then, please, please, please, tell us the perl bug
> > > > number <http://rt.perl.org/perlbug/>, so we can let Red Hat know, so
> > > > they can release a supported update. Help save the world!
> > > >
> > > > _______________________________________________
> > > > LEAD-WebTB mailing list
> > > > address@hidden
> > > > http://www.caps.ou.edu/mailman/listinfo/lead-webtb
> > > >
> > > >
> > > >
> > >
> > > ==========================================================================
> > > =====
> > > Robb Kambic Unidata Program Center
> > > Software Engineer III Univ. Corp for Atmospheric
> > Research
> > > address@hidden WWW: http://www.unidata.ucar.edu/
> > > ==========================================================================
> > > =====
> >
> >
> >
>
> ===============================================================================
> Robb Kambic Unidata Program Center
> Software Engineer III Univ. Corp for Atmospheric Research
> address@hidden WWW: http://www.unidata.ucar.edu/
> ===============================================================================
>